Very Deep Residual Networks with MaxOut for Plant Identification in the Wild

The paper presents our deep learning approach to automatic recognition of plant species from photos. We utilized a very deep 152layer residual network [15] model pre-trained on ImageNet, replaced the original fully connected layer with two randomly initialized fully connected layers connected with maxout [13], and fine-tuned the network on the PlantCLEF 2016 training data. Bagging of 3 networks was used to further improve accuracy. With the proposed approach we scored among the top 3 teams in the PlantCLEF 2016 plant identification challenge.

[1]  Terrance E. Boult,et al.  Towards Open World Recognition , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Hsuan-Tien Lin,et al.  A note on Platt’s probabilistic outputs for support vector machines , 2007, Machine Learning.

[3]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[4]  Andréa Britto Mattos,et al.  Flower Classification for a Citizen Science Mobile App , 2014, ICMR.

[5]  Jiri Matas,et al.  Texture-Based Leaf Identification , 2014, ECCV Workshops.

[6]  Pierre Bonnet,et al.  Plant Identification in an Open-world (LifeCLEF 2016) , 2016, CLEF.

[7]  Kilian Q. Weinberger,et al.  Deep Networks with Stochastic Depth , 2016, ECCV.

[8]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Andrew Zisserman,et al.  Efficient additive kernels via explicit feature maps , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[11]  W. John Kress,et al.  Leafsnap: A Computer Vision System for Automatic Plant Species Identification , 2012, ECCV.

[12]  Song Han,et al.  Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding , 2015, ICLR.

[13]  Andrew Zisserman,et al.  Efficient Additive Kernels via Explicit Feature Maps , 2012, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Itheri Yahiaoui,et al.  A Comparison of Multi-scale Local Binary Pattern Variants for Bark Image Retrieval , 2015, ACIVS.

[15]  Shenghuo Zhu,et al.  Image segmentation for large-scale subcategory flower recognition , 2013, 2013 IEEE Workshop on Applications of Computer Vision (WACV).

[16]  Peter I. Corke,et al.  Content Specific Feature Learning for Fine-Grained Plant Classification , 2015, CLEF.

[17]  Nicu Sebe,et al.  Proceedings of the 21st ACM international conference on Multimedia , 2013, MM 2013.

[18]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[19]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[20]  Terrance E. Boult,et al.  Probability Models for Open Set Recognition , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Hervé Glotin,et al.  LifeCLEF 2016: Multimedia Life Species Identification Challenges , 2016, CLEF.

[22]  Yoshua Bengio,et al.  Maxout Networks , 2013, ICML.

[23]  Nozha Boujemaa,et al.  Pl@ntNet mobile app , 2013, ACM Multimedia.

[24]  R. Sablatnig,et al.  Automated identification of tree species from images of the bark , leaves and needles , 2010 .

[25]  Juan C. Caicedo,et al.  Fine-tuning Deep Convolutional Networks for Plant Recognition , 2015, CLEF.

[26]  Alexis Joly,et al.  LifeCLEF Plant Identification Task 2014 , 2014, CLEF.

[27]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[28]  Sungbin Choi.  Plant Identification with Deep Convolutional Neural Network: SNUMedinfo at LifeCLEF Plant Identification Task 2015 , 2015, CLEF.

[29]  Itheri Yahiaoui,et al.  Pl@ntNet Mobile 2014: Android port and new features , 2014, ICMR.

[30]  Maximilien Servajean,et al.  A Comparative Study of Fine-grained Classification Methods in the Context of the LifeCLEF Plant Identification Challenge 2015 , 2015, CLEF.

[31]  Jiri Matas,et al.  Kernel-mapped histograms of multi-scale LBPs for tree bark recognition , 2013, 2013 28th International Conference on Image and Vision Computing New Zealand (IVCNZ 2013).